In Part I, I discussed how a dice throw isn't as random as you might think, and pondered where randomness really comes from. In the previous Part II, I looked at why randomness is rare (if it exists at all), and asked if the stock market is truly unpredictable or random. This Part III attempts to answer that.
Well, a proponent of the Efficient Market Hypothesis (EMH) would have you believe the stock market is unpredictable. Actually, as I understand the theory, any predictability is arbitraged away by people who do it for no reward whatsoever. This creates a paradox, called the Stiglitz Paradox that I wrote about before:
If a market were informationally efficient, i.e., all relevant information is reflected in market prices, then no single agent would have sufficient incentive to acquire the information on which prices are based.
So if EMH is right, these arbitrageurs do it for free just to tork the rest of us off. So in theory anyway, the market was once predictable, but no longer. Don't look for any EMH proponent to pinpoint that point in time where the market crossed over discontinuously from predictability to unpredictability and every arbitrage that ever will be needed to keep the market efficient was invented. Was it in 1776? 1902? 1929? 1960? There are no answers that I'm aware of, but it doesn't really matter because most EMH believers think the market is a random walk. You need not look for patterns because there are none left. We are before that point in time where the dice throw of the market becomes unpredictable. Arbitrageurs have manufactured randomness from predictability long ago, completely depleting alpha (probably after the first shares traded hands), and have been confounding us with random price movement beta ever since. Capital is allocated with perfect efficiency, all the time, and all that's left to show for it is pure unpredictable randomness.
Again, for those left a little puzzled by the logic of the EMH, the Adaptive Market Hypothesis (AMH) comes to the rescue. In an AMH world, everything has not crossed over fully into randomness (or at least unpredictability) and it probably never will. AMH doesn't say it is easy to predict, however. Arbitageurs and savvy investors thankfully get paid excess returns to wring inefficiencies out of the market, but they miss stuff, leaving some scraps for the next person. This dice throw, in other words, is predictable, if you make an effort at it. CASTrader is my effort at it.
Comments